New Search Engine: Cuil

[Aristotle]

Originally posted by Rosuav
What's been less than impressing you about Google lately?

A lot of worthless results basically. Perhaps more people are gaming their algorithms now, I don't know. But I get less useful results now and have to go through more chaff to find what I want.

But it still stomps this Cuil in the dirt, even for the most basic functionality. These Cuil guys seem to be deliberately ignoring a lot of the things that made google so successful.

[Savaric]

Well it's only been up for a day or two.

Oh... sponsored links. I never realized that was spam, I never payed any attention to them. Damn that, now I'm gonna start noticing it as spam that someone's making money on. God damn, does our entire economy run on advertisements?

[kestra]

Whenever you have to explain the pronunciation and/or meaning of your product, it's an automatic tune out for me.
Seems to me like they overlooked a pretty simple and fundamental marketing principal in their name choice.

[Rosuav]

Originally posted by Savaric
Well it's only been up for a day or two.

Oh... sponsored links. I never realized that was spam, I never payed any attention to them. Damn that, now I'm gonna start noticing it as spam that someone's making money on. God damn, does our entire economy run on advertisements?

It's only been open to the public for a day or two. Obviously they've had their crawler running for ages.... or else they stole Google's database before they left, and never re-crawled.

Does our entire economy run on advertisements? Hmm.... No, but pretty close to. The fact is, advertising does work - or at least, people think it does - so businesses will spend money there to try to gain a competitive advantage. It's possible to run a business without ever spending money on advertising, but in this day and age, it's getting pretty uncommon.

Some companies offer clearly-marked adverts and make no secret of the fact. Others are more subtle, and sell "premium placement" in a listing. But either way, they have to make money. If it's free to get listed and free to view the listings, then someone somewhere has to be paying for it.

Har har har. http://www.cuil.com/info/webmaster_info/ says:

To all those who have contacted us to let us know that they are happy to have their site included in a Web index for the first time, thank you for being a part of the biggest search engine on the Web—Cuil!

How do you measure biggest? Like I said above, Google already have the non-HTML crawling sewn up. Also, a simple look at whois records tells you that Cuil is an upstart but Google know what they're doing... Do a quick whois on the IP address for www.cuil.com and/or any of the crawler IPs listed on the above page, and you'll see they're owned by XO Communications or Performance Systems International. Google's IPs are owned by.... Google. I think that says something about who's bigger.

I'm not going to consider them a force to be reckoned with until I start seeing cuil.com referers in my webalizer stats.

[Gesslar]

I didn't mean to suggest they were SPAM, per se.

Originally posted by Savaric
Well it's only been up for a day or two.

Oh... sponsored links. I never realized that was spam, I never payed any attention to them. Damn that, now I'm gonna start noticing it as spam that someone's making money on. God damn, does our entire economy run on advertisements?

[Aristotle]

Fun article with John C. Dvorak slaughtering Cuil.

[Rosuav]

Originally posted by Aristotle
Fun article with John C. Dvorak slaughtering Cuil.

Unsurprising that Google can find what you're looking for and prioritize the hits ably. That's what they're there for. Surprising, though, that what basically amounts to a Google splinter-group can't get major page results. What have they done? Abandon PageRank? Doubtless they can't use the name nor the database, but you can't copyright/patent the concept of deeming sites that have many links to them as more important than sites with few links. That's a massive oversimplification of PageRank, so if you devise something from the same basic principle and use some different specifics on top of it, I doubt Google's going to be able to sue you for stealing the idea. It'd be like Microsoft trying to claim that they invented the minimize button on the end of your title bar.

Actually, it seems that abandoning PageRank is precisely what they HAVE done.

Rather than rely on superficial popularity metrics, Cuil searches for and ranks pages based on their content and relevance. When we find a page with your keywords, we stay on that page and analyze the rest of its content, its concepts, their inter-relationships and the page’s coherency.

"Superficial"? Granted, it's possible to game the system, spamming links everywhere. But a search engine cannot be authoritative on everything. How can Cuil know what people are likely to look for? Google knows based on what people have previously looked for. (As an aside, even Google's paid ads are ranked by popularity. The ads with the highest click-through rate are shown highest on the page. Google expects its end users - the people doing the searches - will not click on irrelevant ads, and so CTR makes an excellent metric for sorting out the relevant from the useless.)

You know, webmasters have the power to affect whether or not a search engine is of any value. All they need to do is block one or other from crawling the site. As a webmaster, would you feel safe blocking Google? (Assume a significant proportion of your revenue comes from web site traffic.) At the moment, Google has enough marketshare that you'd see a significant drop in traffic if you block it. Cuil? I've yet to see any cuil.com referer entries in my logs, except now when I did a deliberate test. Incidentally, I think Cuil still has not crawled rosuav.com; Google did ages ago, even though I never explicitly submitted it.

A search across all logs for kepl.com.au for the month of August (given that Cuil went public in late July IIRC, going back earlier is pointless) shows the string 'cuil' occurring only in my test entries and in the crawler's UA. So... yeah, it increases our traffic. By the precise load of the crawler, and nothing more.